What is the reason for having a Row Filter node when the Row Splitter node accomplishs the same thing and has the added benifit of not hiding what you filtered out? Is it memory thing?
Just curious.
One is for filtering and the other is for splitting.
I filter if I don’t plan to use the rows that don’t meet my criteria.
I split when I plan on using both outputs, even if it’s just seeing which rows “failed”
1 Like
Hi @rrousselot,
indeed, there are performance differences between the Row Filter and the Row Splitter, due to the fact that the Row Filter can immediately ignore non-matching rows. The Row Splitter on the other hand needs to store all rows, matching and non-matching, as it provides both as outputs. Obviously that difference is larger the more rows are filtered out.
Have a nice day,
nan
3 Likes
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.